Ventana software support forum

If you step through the elements of "priority vector" you will find that the first element has [duration,supplier,ptype] and the next has [duration,supplier,pprofile]. That doesn't work because pprofile is the range. If you replace it with the element ppriority it should work.

It's fairly close to the Numerical Recipes version: http://www.nrbook.com/a/bookcpdf.php Powell's method (and any other hill climber) will find only a single local optimum, unless you do a grid or random variation of the starting point. Then the probability of discovery of additional optima is propo...

It should work as you've described, if the dimensionality of demand is reduced to 1. The demand priority should be greater than the max supply priority, to ensure that demand is always fulfilled. If you want stepwise allocations, the width parameter needs to be narrow.

You can do this temporarily by dragging to zoom in on a regular graph. The setting isn't saved though.

In that case, you don't want the TREND function at all. You want something like a stock of Workforce, with an inflow of hiring or workforce growth or similar, where the inflow (people/month) equals the stock (people) multiplied by a growth rate (fraction per month, e.g. .005).

See also:

Sterman, John D. 1987. Expectation Formation in Behavioral Simulation Models. Behavioral Science, 32:190-211.

Sterman, John D. 2000, Business Dynamics, Boston, MA: McGraw-Hill. Chapter 16.

And:

https://metasd.com/2011/04/delay-sandbox/

There's a sample model for each, attached to the function's topic in the Help system.

You can report the median with the Stats tool.

If you need it dynamically, you store values in an array and use the VECTOR SORT-type functions to find the midpoint.

You are correct that Vensim is usually quantitative, but you can use it for pure diagramming tasks. Here's a video tutorial: https://vensim.com/vensim-causal-loop-diagramming/ Another option is to use Kumu.io. I find it slower for making CLDs, but it does have some nice network analysis features. Tu...

I see. In that case, it does seem sensible to do some kind of combinatorial resampling from the individual columns. Unfortunately, the Vensim file sensitivity isn't set up to do this, but I'll take a look - it might be simple to implement.

Does it make sense to have a small time step in a long-term capacity model? Or if it does, could you then dispense with the array and observe actuals?

Hi - Good question. You are correct, that there isn't a direct way to resample the columns independently in Vensim. You'd have to use R or Python or Excel or something to do that. However.... I think you don't want to resample in that way. If you do, you are in effect sampling each parameter's margi...

When I run this model, T2 starts at 341 and reaches 378 at time 2, and remains there after. T2max is the same, because T2 is always nondecreasing. If I override T2 with a Sin wave of period 8, T2max is still OK - it follows the wave to its maximum and then holds that value. If you mean that T2max at...

This seems to be working fine. T2 and T2max both reach their max values at time 2. What's the problem?

Nope - a vdf output from one model can be read directly as input to another model. No conversion needed - you just need to have matching variable names.

I would recommend importing the spreadsheet to a vdf and using ordinary data variables rather than the GET VDF functions. Then you can control the data source via SIMULATE>DATA| and ADDDATA.

1 & 3 are known and will be fixed in the next release. We'll look into #2.

A possible test would be to compute SUM(Alfa[Sub!]) in a separate variable.

More importantly, using the value of a variable at FINAL TIME to set the value of another stock at INITIAL TIME is acausal, i.e. it implies effect preceding cause. You can't do that in Vensim by design, because no one really knows the future.

I think the thing to do is to discretize the LDC into a number of segments, represented by an array dimension. Then you can compute utilization of each plant type in each segment. Then the price is the highest-cost plant with nonzero utilization in the segment. (The allocation functions might make t...

No problem - we can delete one. Is this the better copy?

Great! At some point, we'll have data entity singletons, so you don't have to bother with creating a key value for something that has only one instance anyway, as here.

Hi - You're really close. The data is actually OK. You can verify this in the results: vresults.png The issue is the references in the Growth entitytype. You have 3: Grass6 GrassDataID GrassGrow (These can be found in the References tab of the Growth inspector.) GrassDataID is the reference correspo...

OK, thanks for the feedback.

The things we'd probably consider, in descending order of priority, would be:
- RK4auto
- RK2/4 fixed
- An implicit method
An Eigenvector method would be cool, but unfortunately not compatible with action scheduling.

At some point, yes, but we haven't had any signals from the user base that it's a priority. What are you looking for?